We have been following-up association signals from our prior 100K genome-wide association study (GWAS) in Pima Indians. Follow-up has included genotyping more than 4200 SNPs in a sample of 3501 full heritage Pima Indians of which 613 SNPs were further genotyped in an independent sample of 3784 mixed heritage Native Americans. We are also comparing our best GWAS signals in Pima Indians with GWAS signals obtained in other populations (predominately Caucasian populations who have submitted their results to public databases). SNPs with the best replication for diabetes associations between the full heritage and mixed heritage samples mapped to KCNQ1 (combined sample P= 2 x 10-6, OR =1.26). Several of these SNPs were previously shown to be associated with diabetes in a GWAS of East Asians. Replication for diabetes associations in full and mixed heritage samples was also observed with SNPs in/near ACAD10 (combined sample P= 5 x 10-5, OR= 1.25). This gene has not been implicated in a GWAS for type 2 diabetes in other ethnic groups. Replication of associations between the full heritage Pima and mixed heritage samples was more common for BMI- associated SNPs as compared to diabetes-associated SNPs perhaps because of the younger age of the mixed heritage sample (median age = 22;therefore many of the non-diabetic individuals will likely develop diabetes). For example we identified common variation in the Map2K3 gene that is strongly associated with BMI in both full heritage and mixed heritage populations. We have found that a large portion of this gene has undergone a duplication but we have shown that only one copy of the gene is functional. We have additionally found variation in the A2BP1 gene that is highly associated with BMI in full heritage Pima Indians and French Caucasian adults. Follow-up studies of our GWAS also identified a variant unique to Pima Indians in the orexin receptor 2 gene that is highly associated with body weight. This gene has a known role in appetite regulation and satiety.